Giernacki, Wojciech (Poznan University of Technology), Kozierski, Piotr (Poznan University of Technology, Institute of Robotics and Machi), Michalski, Jacek (Poznan University of Technology), Retinger, Marek (Poznan University of Poznan), Madonski, Rafal (Jinan University), Campoy, Pascual (Universidad Politecnica Madrid)

Bebop 2 Quadrotor As a Platform for Research and Education in Robotics and Control Engineering

Scheduled for presentation during the Regular Session "UAS Testbeds" (FrD1), Friday, September 4, 2020, 16:50−17:10, Macedonia Hall

2020 International Conference on Unmanned Aircraft Systems (ICUAS), September 1-4, 2020 (Postponed from June 9-12, 2020), Athens, Greece

This information is tentative and subject to change. Compiled on April 18, 2024

Abstract

In conducting research and teaching in fields related to unmanned aerial vehicles (UAVs), it is particularly important to select a universal, safe, open research platform and tools for rapid prototyping. Ready-to-use, low-cost micro-class UAVs such as Bebop 2 are successfully used. This article presents how to use the potential of this flying robot with Robot Operating System (ROS). The most important software solutions for the developed experimental testbed FlyBebop are characterized. Their capabilities in research and education are exemplified here using three distinct cases: 1) research results on the method of optimal, in-flight, iterative self-tuning of UAV position controller parameters (which based only on current measurements), 2) the use of the reinforcement learning method in the autonomous landing of a single drone on a moving vehicle, 3) planning the movement of UAVs for autonomous video recording along the planned path in the arrangement: "cameraman drone" and "lighting technician drones".